annotate toolboxes/FullBNT-1.0.7/nethelp3.3/demard.htm @ 0:cc4b1211e677 tip

initial commit to HG from Changeset: 646 (e263d8a21543) added further path and more save "camirversion.m"
author Daniel Wolff
date Fri, 19 Aug 2016 13:07:06 +0200
parents
children
rev   line source
Daniel@0 1 <html>
Daniel@0 2 <head>
Daniel@0 3 <title>
Daniel@0 4 Netlab Reference Manual demard
Daniel@0 5 </title>
Daniel@0 6 </head>
Daniel@0 7 <body>
Daniel@0 8 <H1> demard
Daniel@0 9 </H1>
Daniel@0 10 <h2>
Daniel@0 11 Purpose
Daniel@0 12 </h2>
Daniel@0 13 Automatic relevance determination using the MLP.
Daniel@0 14
Daniel@0 15 <p><h2>
Daniel@0 16 Synopsis
Daniel@0 17 </h2>
Daniel@0 18 <PRE>
Daniel@0 19 demmlp1</PRE>
Daniel@0 20
Daniel@0 21
Daniel@0 22 <p><h2>
Daniel@0 23 Description
Daniel@0 24 </h2>
Daniel@0 25 This script demonstrates the technique of automatic relevance
Daniel@0 26 determination (ARD) using a synthetic problem having three input
Daniel@0 27 variables: <CODE>x1</CODE> is sampled uniformly from the range (0,1) and has
Daniel@0 28 a low level of added Gaussian noise, <CODE>x2</CODE> is a copy of <CODE>x1</CODE>
Daniel@0 29 with a higher level of added noise, and <CODE>x3</CODE> is sampled randomly
Daniel@0 30 from a Gaussian distribution. The single target variable is determined
Daniel@0 31 by <CODE>sin(2*pi*x1)</CODE> with additive Gaussian noise. Thus <CODE>x1</CODE> is
Daniel@0 32 very relevant for determining the target value, <CODE>x2</CODE> is of some
Daniel@0 33 relevance, while <CODE>x3</CODE> is irrelevant. The prior over weights is
Daniel@0 34 given by the ARD Gaussian prior with a separate hyper-parameter for
Daniel@0 35 the group of weights associated with each input. A multi-layer
Daniel@0 36 perceptron is trained on this data, with re-estimation of the
Daniel@0 37 hyper-parameters using <CODE>evidence</CODE>. The final values for the
Daniel@0 38 hyper-parameters reflect the relative importance of the three inputs.
Daniel@0 39
Daniel@0 40 <p><h2>
Daniel@0 41 See Also
Daniel@0 42 </h2>
Daniel@0 43 <CODE><a href="demmlp1.htm">demmlp1</a></CODE>, <CODE><a href="demev1.htm">demev1</a></CODE>, <CODE><a href="mlp.htm">mlp</a></CODE>, <CODE><a href="evidence.htm">evidence</a></CODE><hr>
Daniel@0 44 <b>Pages:</b>
Daniel@0 45 <a href="index.htm">Index</a>
Daniel@0 46 <hr>
Daniel@0 47 <p>Copyright (c) Ian T Nabney (1996-9)
Daniel@0 48
Daniel@0 49
Daniel@0 50 </body>
Daniel@0 51 </html>